Correlation width changed mean onset-label AUC by −0.012 beyond contact order in 16 SEEG test patients
We tested whether band-limited spectral correlation width adds onset-label information beyond contact order within electrode shafts. We rebuilt spectra from two first-minute public SEEG excerpts for each of 36 patients in OpenNeuro ds004100 v1.1.3. A fixed model used 15 training patients and 16 test patients. Adding width changed mean AUC from 0.8945 to 0.8822. The mean change was −0.0123, with a conditional 95% patient bootstrap interval of −0.0352 to 0.0046. The primary sign-flip p value was 0.375. Twelve test patients already had contact-order AUC 1.0. Width ranks were consistent between files, with median patient correlation 0.8464. Width alone gave mean AUC 0.5584. Adding width to a spectral-shape model changed mean AUC by 0.0005, with an interval that included zero. All eight fixed secondary tests are reported. This known cohort gives no external or clinical outcome validation. The 31 MB package preserves spectra, labels, source code, retrieval pins, results, figures, and reproduction instructions. Independent numerical checks passed. A repeat analysis reproduced all 52 result files.
intracranial EEG · SEEG · epilepsy · spectral correlation width · seizure onset zone · AUC · patient bootstrap
References
5- usesQuadratic log-frequency fits gave lower spectral shape error than straight fits in separate bipolar SEEG files from 36 patients at 20–100 Hz with 55–65 Hz removed sci:sha256:01b8361e8c32258bab97b2e381b2bc4dcc6bb57ea3efc0c8057ad65785c14e7c
This study supplied the 36-patient SEEG cohort, retained contacts, adjacent-pair rules, source metadata, byte retrieval pins, and MIT source code. We retrieved every prefix again and rebuilt broader-band spectra. Its archived 20–100 Hz spectra supplied an additional reconstruction check.
- usesSpectral curvature in fixed models gave a mean onset label AUC gain of 0.027 in 16 SEEG test patients, with an interval that included zero sci:sha256:6801dbc340f4ac049748f756b112da5f28679239577a85dd5527aac41a351dce
This study supplied the clinical flag rule, mismatch exclusions, fixed patient split, and MIT ranking code. We read its outcomes before saving the local width plan. The new test uses width as an added feature and reports this cohort reuse.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
Source dataset for the 36-patient SEEG cohort and the two first-minute recording excerpts per patient.
- doi:10.1093/brain%2Fawad007
- doi:10.3389/fneur.2020.553885
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesBipolar SEEG shape at 20 to 100 Hz gave 55.78 percent expected accuracy for shared fair random shaft labels in 33 patients after target contact exclusion
We used all 36 full pair spectrum archives, contact lists, metadata, raw retrieval pins, and licensed extraction code. Fresh public raw prefixes were used to check these spectra.